ProductionEvidence: Medium65/100

Kingdee Develops Generative AI Features in Two Months on Amazon Bedrock (ERP intelligent ticketing & metrics)

Kingdee International Software Group deployed generative AI features to improve ERP usability and let users retrieve complex data and complete multi-step tasks through natural language. The company built an intelligent ticketing system and an intelligent metrics analysis tool on Amazon Bedrock, integrating agents with Kingdee APIs, knowledge bases, databases, Lambda, API Gateway, and OpenSearch. The system also uses Text2SQL through Bedrock Agents and provides a 24/7 virtual assistant for managers and end users.

Industry
Tech & Comms
Location
China
Published
May 2026

Reported outcomes

100%

accuracyQuality & accuracy

90%accuracy70-80%accuracy

Strategic outcomes

New product / capabilityAdded natural-language ERP task assistanceNew product / capabilityBuilt intelligent ticketing and metrics toolsNew product / capabilityEnabled text-to-SQL ERP data retrievalSpeed & agilityDeployed generative AI features rapidly
Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Accuracy: 100%

AWS Solutions Case StudyMay 27, 2026Customer storyInferred claimMedium evidence strength

The intelligent ticketing system achieved 100 percent accuracy on complex queries in follow-up responses.

Normalized claim

Accuracy: 90%

AWS Solutions Case StudyMay 27, 2026Customer storyInferred claimMedium evidence strength

The metrics tool achieved over 90 percent first-round response accuracy for straightforward queries.

Normalized claim

Accuracy: 70-80%

AWS Solutions Case StudyMay 27, 2026Customer storyInferred claimMedium evidence strength

Complex queries reached 70 to 80 percent accuracy initially before achieving 100 percent accuracy on follow-up responses.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
Kingdee International Software Group
Provider
AWS
Maturity
Production

Kingdee International Software Group deployed generative AI features to improve ERP usability and let users retrieve complex data and complete multi-step tasks through natural language

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 4

  • 1ERP assistant
  • 2Workflow automation
  • 3Decision support
  • Improve ERP usability so users can retrieve complex data and complete multi-step tasks without navigating complex BI reports and interfaces.
  • Streamline difficult ERP workflows such as expense claims and metrics lookup.
  • Support global enterprise operations with secure, natural-language-driven access to ERP data.
  • Built two generative AI solutions on Amazon Bedrock: an intelligent ticketing system and a metrics analysis tool.
  • Used Amazon Bedrock Agents to execute tasks across company systems and data sources, integrating Kingdee APIs and knowledge base.
  • Added a Text2SQL workflow and Amazon OpenSearch Service to convert natural language requests into SQL and retrieve ERP data efficiently.
  • Kingdee deployed the generative AI features within about two months.
  • The intelligent ticketing system achieved 100 percent accuracy on complex queries in follow-up responses.
  • The metrics tool achieved over 90 percent first-round response accuracy for straightforward queries.
  • Complex queries reached 70 to 80 percent accuracy initially before achieving 100 percent accuracy on follow-up responses.
Architecture

Kingdee built intelligent ticketing and metrics analysis solutions on Amazon Bedrock. The implementation uses Amazon Bedrock Agents integrated with Kingdee APIs, databases, AWS Lambda, Amazon API Gateway, and Amazon OpenSearch Service, plus a Text2SQL workflow for natural-language access to ERP data.

Sources & evidence1
Evidence: Medium65/100Evidence strength
  • Customer explicitly identified
  • Deployment status explicitly supported
  • Primary source available
  • Quantified outcome available
  • Technical implementation details available
Type: Customer StoryPublished: May 27, 2026Publisher: AWSEvidence: VendorConfidence: Medium

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Community

Comments

Loading comments...

Similar cases